This journal is© the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 | 29697

Cite this:Phys.Chem.Chem.Phys.,

2015, 17, 29697

On the structure and bonding in the B4O4+

cluster: a boron oxide analogue of the3,5-dehydrophenyl cation with p and r doublearomaticity†

Ting Ou,a Wen-Juan Tian,a Xue-Rui You,a Ying-Jin Wang,ab Kang Wanga andHua-Jin Zhai*ac

Boron oxide clusters offer intriguing molecular models for the electron-deficient system, in which the

boronyl (BO) group plays a key role and the interplay between the localized BO triple bond and the

multicenter electron delocalization dominates the chemical bonding. Here we report the structural,

electronic, and bonding properties of the B4O4+ cationic cluster on the basis of unbiased Coalescence

Kick global-minimum searches and first-principles electronic structural calculations at the B3LYP and

single-point CCSD(T) levels. The B4O4+ cluster is shown to possess a Cs (1, 2A0) global minimum. It

represents the smallest boron oxide species with a hexagonal boroxol (B3O3) ring as the core,

terminated by a boronyl group. Chemical bonding analyses reveal double (p and s) aromaticity in Cs

B4O4+, which closely mimics that in the 3,5-dehydrophenyl cation C6H3

+ (D3h, 1A10), a prototypical

molecule with double aromaticity. Alternative D2h (2, 2B3g) and C2v (3, 2A1) isomeric structures of B4O4+

are also analyzed, which are relevant to the global minima of B4O4 neutral and B4O4� anion,

respectively. These three structural motifs vary drastically in terms of energetics upon changing the

charge state, demonstrating an interesting case in which every electron counts. The calculated ionization

potentials and electron affinities of the three corresponding neutral isomers are highly uneven, which underlie

the conformational changes in the B4O4+/0/� series. The current work presents the smallest boron oxide

species with a boroxol ring, establishes an analogy between boron oxides and the 3,5-dehydrophenyl cation,

and enriches the chemistry of boron oxides and boronyls.

1. Introduction

There has been persistent research interest in boron oxidesduring the past 60 years, which primarily aims at the developmentof highly energetic boron-based propellants.1 Renewed recentinterest in this direction focuses on the electronic, structural,and chemical bonding properties of boron oxide clusters inthe gas phase, where the boronyl (BO) group emerges as a new

inorganic ligand.2 Note that BO and BO� possess a robustBRO triple bond.3,4 These species are isovalent to CN andCN�/CO, respectively, which are well-known inorganic ligands,hinting opportunities to develop the chemistry of boronyls andboron oxides.2,5–13 The geometric structures of boron oxideclusters exhibit vast diversity, which are governed by the com-plexity in their chemical bonding.

Boron is a prototypical electron-deficient element, leadingto unique planar or quasi-planar structures for a wide range ofboron clusters, Bn

� (n = 3–27, 30, 35, 36, 40),14–19 which are instark contrast to bulk boron and boron alloys. The bonding inplanar and quasi-planar boron clusters is governed by multiplearomaticity and antiaromaticity, rendering them interestingall-boron analogues of the aromatic hydrocarbons.20–23 Beyondthe flat world of boron, the first cage-like boron clusters, D2d

B40�/B40, were observed in 2014,19 which are referred to routi-

nely as all-boron fullerenes, or borospherenes. New chiralmembers were subsequently introduced to the borospherenefamily: C3/C2 B39

�, C1 B41+, and C2 B42

2+.24,25 The bonding inborospherenes features the interwoven boron double chains

a Nanocluster Laboratory, Institute of Molecular Science, Shanxi University,

Taiyuan 030006, China. E-mail: hj.zhai@sxu.edu.cnb Department of Chemistry, Xinzhou Teachers University, Xinzhou 034000, Chinac State Key Laboratory of Quantum Optics and Quantum Optics Devices,

Shanxi University, Taiyuan 030006, China

† Electronic supplementary information (ESI) available: Optimized D2h (2B3g) andC2v (2A1) structures of B4O4

+ and C2h (1Ag) structure of B4O4 at the B3LYP/aug-cc-pVTZ level; canonical molecular orbitals (CMOs) of the quasi-linear C2h isomer ofB4O4

+; natural charge distributions of the global minimum and selected low-lyingisomers for B4O4

+: Cs (2A0), C2h (2Bu), D2h (2B3g), and C2v (2A1); and Cartesiancoordinates at the B3LYP/aug-cc-pVTZ level for the four isomeric structures(Cs, C2h, D2h, and C2v). See DOI: 10.1039/c5cp04519c

Received 31st July 2015,Accepted 7th October 2015

DOI: 10.1039/c5cp04519c

www.rsc.org/pccp

PCCP

PAPER

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article OnlineView Journal | View Issue

29698 | Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 This journal is© the Owner Societies 2015

and the p plus s double delocalization, which effectively com-pensates for boron’s intrinsic electron-deficiency.19 This bondingpattern differs fundamentally from that in C60. Boron oxideclusters are the oxidation products of elemental boron clusters.As a consequence, boron oxide clusters are intuitively consideredto be even more electron-deficient. Nonetheless, the available O2p electrons in boron oxide clusters can greatly enrich the BB andBO interactions, leading to exotic chemical bonding. Aromaticityand antiaromaticity, boronyls, multicenter BO rings, and B versusO atomic ratio are some of the key factors that dominate thestructure and bonding in boron oxide clusters.2,26–29

In an early gas-phase experimental work,30 Doyle carried outextensive collision induced dissociation (CID) measurementson a variety of cationic boron oxide clusters, BmOn

+. Amongthese, the B4O4

+ cluster was observed to be unusual, whosecollisional activation spectrum deviates from the general trendof boron oxides. The CID products ‘‘are consistent only with astructure containing a B–B bond’’.30 Based on this observation,Doyle proposed a ‘‘speculative’’ linear structure for B4O4

+, asillustrated in Scheme 1. For a quarter of a century, thisstructure has remained untested. Our recent interest in boronoxide clusters2 has motivated us to revisit this cluster using themodern quantum chemistry tools.

Closely relevant to the current topic, we carried out asystematic computational study lately on small boron oxideclusters, B3On

+/0/� (n = 2–4), which are shown to adopt eitherrhombic or open structures.26 We also predicted the viability ofboronyl boroxine, D3h B6O6, which possesses a boroxol (B3O3)ring as the core with an aromatic p sextet, rendering it a boronoxide analogue of boroxine (B3O3H3) and benzene.27 Further-more, we performed a joint photoelectron spectroscopy (PES)and computational study on the B4O4

0/� clusters,29 establishingthe rhombic and Y-shaped global-minimum structures for B4O4

and B4O4�, respectively. In this contribution, we report on the

structural, electronic, and bonding properties of the cationicB4O4

+ cluster using extensive Coalescence Kick (CK)31,32 struc-tural searches, electronic structural calculations, and canonicalmolecular orbital (CMO) and adaptive natural density partitioning(AdNDP) analyses.33 The B4O4

+ cluster is shown to possess ahexagonal Cs (1, 2A0) global minimum, which marks the exactonset of the boroxol ring in small boron oxide species (that is,the smallest boron oxide cluster with a boroxol ring). Bondinganalyses reveal double (p and s) aromaticity in Cs B4O4

+,rendering it a boron oxide analogue of the 3,5-dehydrophenylcation C6H3

+ (D3h, 1A10).34 The hexagonal Cs B4O4

+ globalminimum is markedly different from the 1988 linear structure(Scheme 1), as well as the rhombic structure of B4O4 and the Y-shaped one of B4O4

�.29 The current Cs B4O4+ global-minimum

structure also helps reinterpret the early experimental collisionalactivation spectrum.30

2. Computational methods

Global-minimum searches for the cationic B4O4+ cluster were

carried out via unbiased CK searches31,32 using the hybridB3LYP method with the small basis set of 3-21G. The structuralsearches were also aided with manual structural constructions.A total of 2000 stationary points on the potential energy surfacewere probed in the CK searches. Low-lying candidate structureswere then fully re-optimized at the B3LYP/aug-cc-pVTZ level,35

which has been established lately to be an appropriate methodfor the boron oxide systems in extensive experimental andcomputational studies.2 Frequency calculations were performedto ensure that all reported structures are true minima. Further-more, single-point CCSD(T) calculations36–38 were done at theB3LYP/aug-cc-pVTZ geometries to benchmark the relative ener-gies for the top low-lying structures. Chemical bonding waselucidated via the AdNDP and CMO analyses. Natural bondorbital (NBO) analysis39 was performed to obtain the naturalatomic charges. The AdNDP calculations were accomplishedusing the AdNDP program33 and all other calculations wereperformed using the Gaussian 09 package.40

3. Results3.1. Global-minimum structure

While one can always be cautious about the capability of modernquantum chemistry in identifying the true global minimum of amolecular system, we are quite confident in the current case fortwo reasons. First, B4O4

+ is a relatively small system and one isexpected to find its global-minimum structure simply via manualstructural constructions, without the need for aid from computa-tional searches. Second, the CK algorithm31,32 is one of the state-of-the-art and well tested methods in unbiased structuralsearches. We reiterate specifically that 2000 stationary pointswere probed on the potential energy surface in our CK searchesfor such a small cluster, which is a heroic effort.

The global minimum of B4O4+, 1 (Cs,

2A0), is depicted inFig. 1(a). Also shown are the bond distances at the B3LYP/aug-cc-pVTZ level. The corresponding symmetric C2v structureof B4O4

+ turns out not to be a true minimum, with oneimaginary frequency of 547.19i cm�1 at the B3LYP level. Tofurther verify that the C2v structure is a first-order saddle point,we have run a complementary PBE0/aug-cc-pVTZ calculation,which also reveals one imaginary frequency (634.15i cm�1). It isthus safe to conclude that the Cs distortion of 1 is not acomputational artifact. As shown in Fig. 2, the structure of 1(Cs,

2A0) is clearly the global minimum of the B4O4+ system, at

both the B3LYP and single-point CCSD(T) levels of theory. It isperfectly planar, being composed of a hexagonal boroxol ringand a terminal boronyl group. The B1B2 and B2O5 distances(1.66 and 1.19 Å) are typical for single B–B and triple BRObonds, respectively,41–43 consistent with the general notion thatthe boronyl group is a robust monovalent s radical.2

The six B–O distances within the boroxol ring in 1 areuneven (Fig. 1(a)), suggesting obvious structural distortions.The two intermediate distances (1.34–1.38 Å) are similar to

Scheme 1 A speculative structure for the B4O4+ cationic cluster,

proposed by Doyle30 in 1988.

Paper PCCP

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article Online

This journal is© the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 | 29699

those in boroxine and boronyl boroxine (1.38 Å at B3LYP/6-311+G(d,p)).27 The two long distances (1.46 Å) appear to beclose to the upper limit of a single B–O bond,41 whereas the twoshort ones (1.28 Å) are markedly shorter than a typical B–Osingle bond and close to a BQO double bond.26,41 Thus, theoverall bonding in the boroxol ring in 1 is substantially greaterthan six B–O single bonds.

3.2. Selected isomeric structures

Alternative low-lying isomeric structures of B4O4+ are shown

in Fig. 2, along with their relative energies at the B3LYP andsingle-point CCSD(T) levels. The energetics at B3LYP andCCSD(T) are generally consistent with each other, suggestingthat the global-minimum structure, 1 (Cs,

2A0), is reasonablywell defined on the potential energy surface. Indeed, thenearest low-lying isomer is 14 and 10 kcal mol�1 above theglobal minimum at B3LYP and CCSD(T) levels, respectively.Among the higher energy isomers, we are interested in thequasi-linear C2h (2Bu), rhombic 2 (D2h, 2B3g), and Y-shaped 3

Fig. 1 Optimized cluster structures at the B3LYP/aug-cc-pVTZ level.(a) Global-minimum structure 1 (Cs,

2A0) of B4O4+. (b) 3,5-Dehydrophenyl

cation (C6H3+). (c) A typical quasi-linear isomeric structure of B4O4

+ on thebasis of a prior proposal of 1988.30 Atoms are labeled numerically and thebond distances are denoted in Å. The B atom is in blue, O in red, C in thickgray, and H in light gray.

Fig. 2 Alternative optimized isomeric structures of B4O4+ at the B3LYP/aug-cc-pVTZ level. The relative energies (in kcal mol�1) are indicated at the

B3LYP/aug-cc-pVTZ and the single-point CCSD(T)//B3LYP/aug-cc-pVTZ (in italic) levels. The B atom is in blue and O is in red.

PCCP Paper

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article Online

29700 | Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 This journal is© the Owner Societies 2015

(C2v, 2A1), which are approximately 29, 88, and 82 kcal mol�1

above the global minimum 1. All the three structures are alsoperfectly planar.

Quasi-linear C2h (2Bu) is illustrated in Fig. 1(c). It can beconsidered as the fusion of two BOBO units, or four BO units.The two terminal BO groups possess bond distances of 1.19 Åand are readily assigned as boronyls.2 The remaining two BOunits have distances of 1.26 Å, which also show triple bondcharacter but are slightly longer than typical BRO bonds.2,4

The B2O7 and B1O8 bonds (1.35 Å) are roughly B–O singlebonds, whereas the B3B4 distance (1.59 Å) is between typicalB–B single and BQB double bonds.12,41 Note that this isomer isbasically in the spirit of the ‘‘speculative’’ linear structure ofB4O4

+, proposed initially in 1988.30 However, the bondingdetails in the two structures differ substantially (Fig. 1(c) versusScheme 1).

Isomer 2 (D2h, 2B3g) possesses a rhombic B2O2 core, attachedby two terminal boronyl groups (Fig. 2). The terminal BO andBB distances (1.24 and 1.69 Å, respectively; Fig. S1 in the ESI†)are consistent with triple BRO and single B–B bonds,2,41

whereas the rhombic BO distances (1.40 Å) are comparable toa B–O single bond. A similar D2h (1Ag) structure was recentlylocated as the global minimum for the B4O4 neutral cluster,29

whose BRO, B–B, and B–O distances are 1.20, 1.67, and 1.40 Å,respectively. Only minor structural differences are observedbetween 2 of B4O4

+ and D2h (1Ag) of B4O4.The Y-shaped isomer 3 (C2v, 2A1) of B4O4

+ (Fig. 2 and Fig. S1in the ESI†) are associated with the Cs (2A00) global minimum ofthe B4O4

� anion cluster.29 In the latter, the terminal OBO isbent. The two terminal BO groups in 3 (1.19 Å) are boronyls.The short and long BO distances in the OBO unit (1.24 versus1.29 Å) are approximately assigned as BRO and BQO bonds,respectively.2,41 The BO distance associated with the central Batom (1.34 Å) is a B–O single bond. The two BB distances in 3(1.70 Å) are slightly elongated with respect to the B–B singlebond in 1 (1.66 Å; Fig. 1(a)), consistent with the fact that only3 electrons are available for the two BB bonds in 3 (due to thepositive charge). The formal BB bond order is 0.75. Going from3 of B4O4

+ to the Cs (2A00) global minimum of B4O4�, one

observes bond distance changes associated with the center Batom: the B–O and B–B bonds are 1.46 and 1.64 Å, respectively,in the latter structure.29

4. Discussion4.1. Hexagonal Cs B4O4

+ cluster: the genesis of the boroxolring in boron oxides

Bulk boron oxides are amorphous and difficult to crystallize,whose structures have remained controversial in the literature.However, it is generally believed that the boroxol ring consti-tutes a large fraction of them, probably up to 75%.44 Further-more, it has been speculated that the high temperature B2O3

liquids are also networked by the boroxol rings, terminatedwith boronyl groups.45 It is thus desirable to characterize theintrinsic stability and bonding of the boroxol ring in gas-phase

boron oxide clusters. In particular, what is the smallest,free-standing boron oxide cluster that can support a boroxolring? In other words, how many B/O atoms are needed at theminimum in order to self-assemble into a chemical species thatcontains a boroxol ring? The present results offer a definiteanswer.

The lower bound to explore the onset of the boroxol ring isthe B3On

�/0/+ (n Z 3) cluster. However, our recent systematiccomputational searches on a series of B3On

�/0/+ (n = 2–4)clusters concluded that the boroxol ring is not present inthem, and only three out of these nine species have a rhombicB2O2 ring.26 As an upper bound, the boronyl boroxine (B6O6)cluster27 was shown to possess a robust boroxol ring already,rendering it an analogue of benzene and boroxine. Betweenthe lower and upper bounds, we performed combined experi-mental and computational studies on the B4O3

�/0 and B4O4�/0

clusters,11,29 which are either open structures or possess arhombic B2O2 ring. The present B4O4

+ (1, Cs,2A0) cluster thus

marks the exact onset of the boroxol ring in boron oxides,which represents the smallest boron oxide species with aboroxol ring. We stress that the genesis of the boroxol ring inB4O4

+ is governed by both geometric and electronic reasons; seeSections 4.2 and 4.3.

4.2. Double (p and r) aromaticity: Cs B4O4+ versus D3h C6H3

+

The 1 (Cs,2A0) global-minimum structure of the B4O4

+ clusterexhibits perfectly planar, annular shape, whose six-memberedboroxol ring is known to possess a p sextet,27 akin to benzeneand boroxine. To understand the bonding nature of B4O4

+

(1, Cs,2A0), we performed detailed CMO and AdNDP33 analyses.

Key CMOs relevant to the delocalized bonding in the B3O3 ringof 1 are shown in Fig. 3(a). The combination of HOMO�5(HOMO denotes the highest occupied molecular orbital),HOMO�7, and HOMO�9 constitutes a delocalized 6p system,which is typical for complexes containing a boroxol ring,27

rendering p aromaticity for 1 according to the (4n + 2) Huckelrule. This 6p system originates from three O 2p lone-pairs, withthe O6, O7, and O8 atoms each contributing one.46 The AdNDPdata fully recover the 6p system (Fig. 4(a); third row), which ispresented as three three-center two-electron (3c–2e) BOBbonds, that is, the Kekule structure. Note that the six B–O

Fig. 3 Comparison of the delocalized p and s canonical molecularorbitals (CMOs) of (a) Cs B4O4

+ (1) and (b) D3h C6H3+. SOMO stands for

the single occupied molecular orbital.

Paper PCCP

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article Online

This journal is© the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 | 29701

s bonds in the ring, the B–B s bond, and the terminal BROtriple bond are also recovered in the AdNDP results, as antici-pated (Fig. 4(a); the first and second rows).

It is interesting to note the singly occupied molecular orbital(SOMO; Fig. 3(a)) of 1. This is a s CMO, being composed ofB 2s/2p atomic orbitals (AOs) from B3, B4, and B1. While thecontributions of the three B centers are not even, the SOMO istruly 3c–1e in nature, which is completely bonding. In theAdNDP analysis, the 3c–1e s bond is also fully recovered(Fig. 4(a); third row). A delocalized 3c–1e s bond is known tobe s aromatic, similar to a delocalized 3c–2e s system, whichfollows the (4n + 2) Huckel rule for s aromaticity. Thus, 1(Cs,

2A0) may be classified to be s aromatic. Note that the 3c–1es bond is only half-occupied, which is relatively weak due to thelarge BB distances in the boroxol ring. To be specific, the B3B4distance is 2.176 Å, which is far beyond the recommendedupper limit for a B–B single bond (1.70 Å).41 Consequently, theresonance stabilization of s aromaticity in 1 is only half withrespect to a 3c–2e s aromatic system. This explains, at leastpartially, why structure 1 as the global minimum of the B4O4

+

cationic system adopts a distorted Cs structure, rather than C2v.The delocalized p and s frameworks of 1 (Fig. 3(a) and 4(a))

collectively define a doubly (p and s) aromatic system, whoseapproximate Lewis presentation is illustrated in Fig. 5(a).To the best of our knowledge, s aromaticity has not beendiscussed for binary BO ring molecular systems to date. Double(p and s) aromaticity in boron oxides has remained unknown

in the literature. The B4O4+ (1, Cs,

2A0) cluster is thus a uniquespecies. It must be stressed that chemical bonding in boronoxides involves the close interplay of robust BRO triple bondsversus B–O/BQO bonds, localized versus delocalized bonding,cyclic versus open structures, rhombic versus hexagonal rings,etc. For such electron-deficient systems, it is rather challengingto simultaneously reach the ideal B/O atomic ratio as well as thecorrect electron counting. The only way toward the discovery ofa double aromatic boron oxide species is to systematically tunethe cluster size, the B/O ratio, and the charge state. In thisprocess, every B/O atom makes a difference and every electroncounts.

Double (p and s) aromaticity in B4O4+ (1, Cs,

2A0) is immediatereminiscent of the 3,5-dehydrophenyl cation molecule, C6H3

+

(D3h, 1A10) (Fig. 1(b)).34 The latter species is often quoted as

one of the milestone studies in the development of aromaticityconcept via computational design. Indeed, the delocalized p ands CMOs of C6H3

+ (Fig. 3(b)) show one-to-one correspondence tothose of B4O4

+ (1, Cs,2A0), except that the s CMO in 1 is distorted

due to its Cs symmetry. This similarity in the bonding patternsuggests that the B4O4

+ (1, Cs,2A0) cluster is a boron oxide

analogue of the 3,5-dehydrophenyl cation. The AdNDP data alsosupport the remarkable analogy between the two species (Fig. 4),whose p systems are both illustrated as the Kekule structures. Itis important to note that the structural distortion of B4O4

+ (1) toCs symmetry results in shortened B4O6/B4O8 (1.28 Å) versuselongated B1O8/B3O6 (1.46 Å) bonds. The former distance isclose to a BQO bond,2,41 which reflects the collective effects ofthe localized 2c–2e B–O s bond, as well as the delocalized p sextetand the 3c–1e s bond. The unevenness in bond distances is anindication that the p sextet is not uniformly distributed in theB3O3 ring, but it not necessarily suggests a localized 2c–2e p bondin B4O6 or B4O8.

The isolobal analogy between B4O4+ (1) and C6H3

+ can alsobe revealed from a simple valence electron counting in the B3O3

versus C6 hexagonal rings (Fig. 1). The C6 ring in C6H3+ has 23

valence electrons including the positive charge; the latter isdistributed on the C2/C4/C6 centers. Three of these electronsare used for the C–H bonds, resulting in 20 electrons forbonding within the C6 ring. Of these, 12 form the six C–C

Fig. 4 The AdNDP bonding patterns for (a) B4O4+ (1) and (b) C6H3

+. Thefour O 2s lone pairs in 1 are not depicted. The occupation numbers (ONs)are shown.

Fig. 5 Schematic Lewis presentations of (a) the hexagonal Cs (1) globalminimum and (b) the quasi-linear C2h isomer of B4O4

+. The delocalizedbonding in (a) features a p sextet (in red) and a 3c–1e s bond (in blue;dashed lines), rendering double aromaticity in 1. The quasi-linear isomer(b) possesses dual 3c–4e B–O–B p bonds, shown as dashed lines in red,where a dashed line corresponds effectively to a bond order of 0.5.

PCCP Paper

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article Online

29702 | Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 This journal is© the Owner Societies 2015

2e–2e s bonds. The remaining 8 electrons are responsible forglobal delocalized bonding (6p plus 2s; Fig. 3(b)). Here everyC center contributes one electron to the p sextet. Similarly,the B3O3 ring in B4O4

+ (1) has 20 valence electrons, wherethe positive charge is counted and three O 2s lone-pairs arenot counted. Of these, one B center uses an electron for theterminal B–(BO) single bond and all B centers use six electronsin total for the six B–O 2e–2e s bonds, whereas the threeO centers provide the remaining six electrons for the latterbonds. Thus only seven electrons are left for global, delocalizedbonding in the B3O3 ring: six electrons from the O centers formthe p sextet and the single electron from the B centers makesthe 3c–1e s bond.

Note that the terminal BO group in B4O4+ (1) is isolobal to an

H terminal in C6H3+, because both are monovalent s radicals.

The difference between the B3O3 and C6 hexagonal rings is thatthe C1/C3/C5 centers in the latter, which form terminal C–Hbonds (Fig. 1(b)), are relatively electron poor and thus do notparticipate in s aromaticity. In contrast, the B centers in B4O4

+

(1), one of which also has a terminal ligand (Fig. 1(a)), manageto reserve one electron for the 3c–1e s bond, despite the factthat the O centers have more valence electrons. This is becausethe O centers are solely responsible for the p sextet, with theB centers only offering the corresponding AOs and not the electrons.Therefore, in terms of the bonding analogy, the B1/B3/B4centers in B4O4

+ (1) (Fig. 1(a)) should correspond to theC6/C2/C4 centers in C6H3

+ (Fig. 1(b)). The former centersin B4O4

+ (1) have relatively fewer valence electrons and areassociated with a terminal BO group, whereas the lattercenters in C6H3

+ are relatively electron-rich and do not bindto the H terminals. The above correspondence is exactlyreflected in the orientation of the delocalized p and s CMOs,as depicted in Fig. 3.

Double aromaticity in B4O4+ (1, Cs,

2A0) may be furthercharacterized using the nucleus-independent chemical shift(NICS). The calculated NICS values at the B3LYP level at thehexagon ‘‘center’’ and at 1 Å above it are NICS(0) = �5.6 ppmand NICS(1) = �3.4 ppm, where the ‘‘center’’ is defined as thearithmetic mean of Cartesian coordinates of the six B/O atomsin the hexagonal ring. Both of the NICS values are negative,confirming that 1 indeed possesses (p and s) double aromati-city. For comparison, those calculated at the same level forD3h C6H3

+ are NICS(0) = �43.8 ppm and NICS(1) = �19.7 ppm.It is stressed that boron oxide clusters always have weakeraromaticity with respect to typical hydrocarbons, because thep aromaticity in the former species originates from O 2p lone-pairs, whose extent of delocalization is relatively poor.47 Forinstance, a recent report gives a NICS value of �3.4 ppm for theprototypical ‘‘inorganic benzene’’, boroxine B3O3H3, at theB3LYP level, which is in contrast to �29.7 ppm for benzene.27

One reviewer suggests an interesting possibility of exploringsimilar bonding phenomena in the relevant Si–O systems or inclusters with CO ligands. While this is clearly beyond the scopeof the current work, we would like to comment here that it isdoable through careful computational design, by tuning thesize, composition, and charge state. Nonetheless, considering

the large difference in electronegativity between Si and O, 1.90for Si versus 3.44 for O at the Pauling scale, the electrons (both pand s) can be far more difficult to delocalize in a Si–O ring.Aromaticity in such molecules should thus be rather weak.

4.3. B4O4+/0/�: dependence of conformation on the charge

state

Very recently, we reported the global-minimum structures ofB4O4

� and B4O4 clusters via anion PES experiments, globalstructural searches, and electronic structure calculations.29 Asshown in Fig. 6(b), the global-minimum structures of B4O4

+/0/�

are distinctly different: hexagonal B4O4+ (1, Cs,

2A0), rhombicB4O4 (4, D2h, 1Ag), and Y-shaped B4O4

� (7, Cs,2A00). All three

types of conformation for three charge states (cation, neutral,and anion) were optimized herein at the B3LYP/aug-cc-pVTZlevel, and the relative energies for the nine structures are shownin Fig. 6(b). The relative energies were also refined at the single-point CCSD(T) level. Notably, the stability for a specific con-formation is sensitively charge-state dependent. For example,the hexagonal isomer is the global minimum for B4O4

+ (1);whereas it becomes the least stable of the three for B4O4 (6) orB4O4

� (9), being 34 and 55 kcal mol�1 above the globalminimum at the CCSD(T) level, respectively. The Y-shapedisomer is 82 kcal mol�1 above the global minimum for B4O4

+

(3). However, it gradually gains stability with the addition ofextra electrons, which is only 18 kcal mol�1 above the globalminimum for B4O4 (5) and further becomes the global minimumfor B4O4

� (7).To be quantitative, going from B4O4

+ to B4O4 (Fig. 6(b)), therhombic and Y-shaped isomers gain an enhanced stability of122 kcal mol�1 (2 - 4) and 98 kcal mol�1 (3 - 5) with respectto the hexagonal isomer (1 - 6). Similarly, going from B4O4 toB4O4

�, the Y-shaped isomers gain an enhanced stability of 26kcal mol�1 (5 - 7) with respect to the rhombic isomer (4 - 8).The above enhancement of stability is remarkable, demonstrat-ing an interesting electronic system in which a single electroncan make a difference. This observation may be entirely attri-buted to boron’s intrinsic electron-deficiency.

As an alternative presentation, the dependence of conforma-tion on the charge state in the B4O4

+/0/� series can be rationa-lized using the ionization potentials (IPs) and electron affinities(EAs) for the three isomeric structures of B4O4 (4–6; Fig. 6(a)),which are calculated at both the B3LYP/aug-cc-pVTZ and single-point CCSD(T) levels. The IP of the hexagonal isomer B4O4 (6) is8.46 eV at CCSD(T), being substantially lower than those of therhombic and Y-shaped isomers (13.75 eV for 4; 12.70 eV for 5).The extremely low IP of 6 manages to gain an enhanced stabilityof 4.2–5.3 eV for the hexagonal isomer in the cationic chargestate, which offsets the relative energy of 6 with respect to 4/5.This explains why 1 becomes the global minimum for B4O4

+.On the other hand, Y-shaped 5 possesses a much higher EA(2.57 eV) than other isomers (1.43 eV for 4; 0.89 eV for 6), whichmakes the Y-shaped isomer (7) the global minimum for theanion. The low EA of hexagonal isomer 6 further reinforces theinstability of hexagonal isomer 9 as an anion.

Paper PCCP

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article Online

This journal is© the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 | 29703

Not surprisingly, the calculated IP and EA values are closelyrelated to the nature of the frontier CMOs of 4–6 (Fig. 7). TheHOMOs (Fig. 7(a)) determine the IPs of 4–6. It is nonbondingbased on O 2p AOs for 4, BO p bonding for 5, and B based for 6,consistent with their decreasing IP values. In particular, the IPvalue drops sharply from 4/5 to 6. For the lowest unoccupiedmolecular orbitals (LUMOs; Fig. 7(b)), it is p bonding based onall B centers but with partial antibonding character for 4,

completely p bonding based on three B centers for 5, and Bbased s CMO with antibonding character for 6, which areconsistent with the fact that 5 has the highest EA and 6 hasthe lowest. In summary, the IPs and EAs of the three neutralconformers (4–6) appear to hold the key to the structuralchanges along the B4O4

+/0/� series.It is interesting to comment here that, despite its bonding

nature of double aromaticity, the hexagonal structure is theglobal minimum for the B4O4

+ (1) cationic cluster only (Fig. 6).With one or two extra electrons being added to the system, therhombic or Y-shaped structures are more favorable, becausethe extra electrons enter distinctly different orbitals in differentisomers (Fig. 7). Thus the doubly aromatic B4O4

+ (1) clusteras the global minimum is a delicate balance of the B/O atomicratio, atomic connectivity, and electron counting. In this respect,‘‘aromaticity’’ does not necessarily dominate the stability of thesystem.

4.4. On the ‘‘speculative’’ structure of 1988: quasi-linearC2h B4O4

+

One of the motivations of this computational study on theB4O4

+ cluster comes from the 1988 experimental work byDoyle,30 who examined a variety of BmOn

+ using the CIDtechnique. The collisional activation spectrum of the B4O4

+

cluster deviates from the general trend of BmOn+ dissociation,

Fig. 6 (a) Schematic diagram (not to scale) showing the relative energies of three isomeric structures in three charge states. The calculated ionizationpotentials (IPs) and electron affinities (EAs) are shown in eV at the B3LYP and single-point CCSD(T) (in parentheses) levels. (b) Structures of 1–9 and theirrelative energies (in kcal mol�1) at the B3LYP and single-point CCSD(T) (in parentheses) levels. The B atom is in blue and O is in red.

Fig. 7 Pictures of (a) the highest occupied molecular orbitals (HOMOs)and (b) the lowest unoccupied molecular orbitals (LUMOs) for the threetypical isomeric structures of B4O4 (4–6). These orbitals hold the key tounderstanding the structural evolution of B4O4

+/0/� with charge states(Fig. 6).

PCCP Paper

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article Online

29704 | Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 This journal is© the Owner Societies 2015

suggesting the presence of a BB bond in B4O4+. Doyle thus

proposed a ‘‘speculative’’ linear structure for B4O4+ (Scheme 1).

Our current study appears to be the first to systematicallyexamine the potential energy surface of B4O4

+ via global struc-tural searches (Fig. 2). The quasi-linear C2h (2Bu) structure(Fig. 1(c)) should best match that proposed by Doyle, althoughtheir bonding details differ markedly. The C2h (2Bu) structureturns out to be B29 kcal mol�1 above the global minimum atthe single-point CCSD(T) level (Fig. 2), which was thus unlikelyto be present in the 1988 experiments.

Despite its high energetics, chemical bonding in the quasi-linear B4O4

+ C2h (2Bu) structure remains interesting. Fig. 8shows a few key CMOs. HOMO�11 is the B–B s bond, whereasthe p SOMO is also primarily responsible for the B–B bonding.These collectively define a B–B bond with a formal order of 1.5,consistent with the relatively short distance of 1.59 Å, which fallbetween typical B–B and BQB bonds.12,41 HOMO�9/HOMO�7represents a pair of 3c–2e B–O–B p bonds (B2–O7–B3 andB4–O8–B1). Similarly, HOMO�10/HOMO�8 represents anotherpair of 3c–2e B–O–B p bonds. Other CMOs are shown in Fig. S2in the ESI.† Overall, the bonding in C2h (2Bu) can be summar-ized as the Lewis presentation, which suggests that the B2O5,B2O7, and B3O7 bonds have very different bond orders (that is,effectively three, one, and two, respectively; Fig. 5(b)), consis-tent with their bond distances of 1.19, 1.35, and 1.26 Å.2,41 Westress that this Lewis presentation differs markedly from the1988 proposal (Scheme 1).30

Based on the current data, it is imperative that we reinterpretthe experimental CID data using the global-minimum structure:B4O4

+ (1, Cs,2A0). The observed CID products of B4O4

+ are B2O2+

(loss of B2O2) and B3O3+ (loss of BO).30 In 1 (Cs,

2A0), the positivecharge is located on the B4 and B3 centers, in particular on B4(Fig. 1(a) and Fig. S3(a) in the ESI†). The cleavage of B1O8 andB3O7 bonds will produce B2O2

+ with the loss of B2O2, whereas thecleavage of the B1B2 bond will produce B3O3

+ with the loss of BO.These are the possible dissociation channels observed. Interest-ingly, the latter channel hints that a hexagonal B3O3

+ ring mightbe accessible as a CID product, although the gas-phase B3O3

+

cluster possesses a linear global-minimum structure.26

5. Conclusions

In this paper, we report on the first-principles theoreticalprediction of the global-minimum structure of the boron oxideB4O4

+ cluster. The calculations include extensive Coalescence

Kick structural searches, electronic structural calculations atthe B3LYP and single-point CCSD(T) levels, and chemicalbonding analyses. The B4O4

+ cationic cluster is shown to possessa perfectly planar, hexagonal (Cs,

2A0) structure as the globalminimum, which contains a boroxol B3O3 ring and a terminalboronyl group. It marks the exact onset of the boroxol ring inboron oxide clusters. Chemical bonding analyses show that B4O4

+

(Cs,2A0) features double (p and s) aromaticity, rendering it an

interesting boron oxide analogue of the 3,5-dehydrophenylcation. The hexagonal (Cs,

2A0) global-minimum structure ofB4O4

+ differs marked from a ‘‘speculative’’ linear structureproposed in 1988, as well as from the recently establishedrhombic and Y-shaped global minima for B4O4 and B4O4

�,respectively. The B4O4

+/0/� series represent a unique molecularsystem, in which a single electron can make a difference. TheB4O4

+ (Cs,2A0) global-minimum structure also helps reinterpret

the experimental collisional activation spectrum in the literature.

Acknowledgements

This work was supported by the National Natural ScienceFoundation of China (21573138) and in part by the State KeyLaboratory of Quantum Optics and Quantum Optics Devices(KF201402). H.J.Z. gratefully acknowledges the start-up fundfrom Shanxi University for support.

References

1 S. H. Bauer, Chem. Rev., 1996, 96, 1907.2 H. J. Zhai, Q. Chen, H. Bai, S. D. Li and L. S. Wang,

Acc. Chem. Res., 2014, 47, 2435.3 P. G. Wenthold, J. B. Kim, K. L. Jonas and W. C. Lineberger,

J. Phys. Chem. A, 1997, 101, 4472.4 H. J. Zhai, L. M. Wang, S. D. Li and L. S. Wang, J. Phys. Chem.

A, 2007, 111, 1030.5 D. Yu. Zubarev, A. I. Boldyrev, J. Li, H. J. Zhai and L. S.

Wang, J. Phys. Chem. A, 2007, 111, 1648.6 H. Bai, H. J. Zhai, S. D. Li and L. S. Wang, Phys. Chem. Chem.

Phys., 2013, 15, 9646.7 H. Bock, L. Cederbaum, W. von Niessen, P. Paetzold,

P. Rosmus and B. Solouki, Angew. Chem., Int. Ed. Engl.,1989, 28, 88.

8 M. L. Drummond, V. Meunier and B. G. Sumpter, J. Phys.Chem. A, 2007, 111, 6539.

9 (a) T. B. Tai and M. T. Nguyen, Chem. Phys. Lett., 2009,483, 35; (b) M. T. Nguyen, M. H. Matus, V. T. Ngan, D. J.Grant and D. A. Dixon, J. Phys. Chem. A, 2009, 113, 4895;(c) T. B. Tai, M. T. Nguyen and D. A. Dixon, J. Phys. Chem. A,2010, 114, 2893.

10 C. B. Shao, L. Jin and Y. H. Ding, J. Comput. Chem., 2011,32, 771.

11 H. J. Zhai, S. D. Li and L. S. Wang, J. Am. Chem. Soc., 2007,129, 9254.

12 S. D. Li, H. J. Zhai and L. S. Wang, J. Am. Chem. Soc., 2008,130, 2573.

Fig. 8 Selected canonical molecular orbitals (CMOs) of the quasi-linearB4O4

+ C2h (2Bu) isomer. SOMO stands for the single occupied molecularorbital.

Paper PCCP

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article Online

This journal is© the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 | 29705

13 (a) H. J. Zhai, J. C. Guo, S. D. Li and L. S. Wang, Chem-PhysChem, 2011, 12, 2549; (b) H. Bai, H. J. Zhai, S. D. Li andL. S. Wang, Phys. Chem. Chem. Phys., 2013, 15, 9646;(c) Q. Chen, H. J. Zhai, S. D. Li and L. S. Wang, J. Chem.Phys., 2012, 137, 044307; (d) Q. Chen, H. Bai, H. J. Zhai,S. D. Li and L. S. Wang, J. Chem. Phys., 2013, 139, 044308;(e) H. J. Zhai, C. Q. Miao, S. D. Li and L. S. Wang, J. Phys.Chem. A, 2010, 114, 12155; ( f ) W. Z. Yao, J. C. Guo, H. G. Luand S. D. Li, J. Phys. Chem. A, 2009, 113, 2561; (g) H. J. Zhai,Q. Chen, H. Bai, H. G. Lu, W. L. Li, S. D. Li and L. S. Wang,J. Chem. Phys., 2013, 139, 174301; (h) D. Z. Li, L. J. Zhang,T. Ou, H. X. Zhang, L. Pei, H. J. Zhai and S. D. Li, Phys.Chem. Chem. Phys., 2015, 17, 16798; (i) W. J. Tian, H. G. Xu,X. Y. Kong, Q. Chen, W. J. Zheng, H. J. Zhai and S. D. Li,Phys. Chem. Chem. Phys., 2014, 16, 5129.

14 (a) H. J. Zhai, L. S. Wang, A. N. Alexandrova and A. I.Boldyrev, J. Phys. Chem. A, 2003, 107, 9319; (b) A. N.Alexandrova, A. I. Boldyrev, H. J. Zhai, L. S. Wang,E. Steiner and P. W. Fowler, J. Phys. Chem. A, 2003,107, 1359; (c) A. N. Alexandrova, A. I. Boldyrev, H. J. Zhaiand L. S. Wang, J. Phys. Chem. A, 2004, 108, 3509; (d) H. J.Zhai, A. N. Alexandrova, K. A. Birch, A. I. Boldyrev andL. S. Wang, Angew. Chem., Int. Ed., 2003, 42, 6004; (e) A. N.Alexandrova, H. J. Zhai, L. S. Wang and A. I. Boldyrev, Inorg.Chem., 2004, 43, 3552.

15 (a) A. P. Sergeeva, D. Yu. Zubarev, H. J. Zhai, A. I. Boldyrevand L. S. Wang, J. Am. Chem. Soc., 2008, 130, 7244;(b) C. Romanescu, D. J. Harding, A. Fielicke and L. S.Wang, J. Chem. Phys., 2012, 137, 014317.

16 A. N. Alexandrova, A. I. Boldyrev, H. J. Zhai and L. S. Wang,Coord. Chem. Rev., 2006, 250, 2811.

17 (a) B. Kiran, S. Bulusu, H. J. Zhai, S. Yoo, X. C. Zeng andL. S. Wang, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 961;(b) Z. A. Piazza, W. L. Li, C. Romanescu, A. P. Sergeeva,L. S. Wang and A. I. Boldyrev, J. Chem. Phys., 2012, 136,104310; (c) A. P. Sergeeva, Z. A. Piazza, C. Romanescu,W. L. Li, A. I. Boldyrev and L. S. Wang, J. Am. Chem. Soc.,2012, 134, 18065; (d) I. A. Popov, Z. A. Piazza, W. L. Li,L. S. Wang and A. I. Boldyrev, J. Chem. Phys., 2013,139, 144307; (e) Z. A. Piazza, I. A. Popov, W. L. Li, R. Pal,X. C. Zeng, A. I. Boldyrev and L. S. Wang, J. Chem. Phys.,2014, 141, 034303.

18 (a) W. L. Li, Y. F. Zhao, H. S. Hu, J. Li and L. S. Wang, Angew.Chem., Int. Ed., 2014, 53, 5540; (b) W. L. Li, Q. Chen,W. J. Tian, H. Bai, Y. F. Zhao, H. S. Hu, J. Li, H. J. Zhai,S. D. Li and L. S. Wang, J. Am. Chem. Soc., 2014, 136, 12257;(c) Z. A. Piazza, H. S. Hu, W. L. Li, Y. F. Zhao, J. Li andL. S. Wang, Nat. Commun., 2014, 5, 3113; (d) Q. Chen,G. F. Wei, W. J. Tian, H. Bai, Z. P. Liu, H. J. Zhai andS. D. Li, Phys. Chem. Chem. Phys., 2014, 16, 18282.

19 H. J. Zhai, Y. F. Zhao, W. L. Li, Q. Chen, H. Bai, H. S. Hu,Z. A. Piazza, W. J. Tian, H. G. Lu, Y. B. Wu, Y. W. Mu,G. F. Wei, Z. P. Liu, J. Li, S. D. Li and L. S. Wang, Nat. Chem.,2014, 6, 727.

20 H. J. Zhai, B. Kiran, J. Li and L. S. Wang, Nat. Mater., 2003,2, 827.

21 W. Huang, A. P. Sergeeva, H. J. Zhai, B. B. Averkiev, L. S.Wang and A. I. Boldyrev, Nat. Chem., 2010, 2, 202.

22 J. E. Fowler and J. M. Ugalde, J. Phys. Chem. A, 2000,104, 397.

23 J. I. Aihara, H. Kanno and T. Ishida, J. Am. Chem. Soc., 2005,127, 13324.

24 Q. Chen, W. L. Li, Y. F. Zhao, S. Y. Zhang, H. S. Hu, H. Bai,H. R. Li, W. J. Tian, H. G. Lu, H. J. Zhai, S. D. Li, J. Li andL. S. Wang, ACS Nano, 2015, 9, 754.

25 Q. Chen, S. Y. Zhang, H. Bai, W. J. Tian, T. Gao, H. R. Li,C. Q. Miao, Y. W. Mu, H. G. Lu, H. J. Zhai and S. D. Li,Angew. Chem., Int. Ed., 2015, 54, 8160.

26 Q. Chen, H. G. Lu, H. J. Zhai and S. D. Li, Phys. Chem. Chem.Phys., 2014, 16, 7274.

27 D. Z. Li, H. Bai, Q. Chen, H. G. Lu, H. J. Zhai and S. D. Li,J. Chem. Phys., 2013, 138, 244304.

28 W. Wang, Q. Chen, Y. J. Wang, H. Bai, T. T. Gao, H. R. Li,H. J. Zhai and S. D. Li, Phys. Chem. Chem. Phys., 2015,17, 19929.

29 W. J. Tian, L. J. Zhao, Q. Chen, T. Ou, H. G. Xu, W. J.Zheng, H. J. Zhai and S. D. Li, J. Chem. Phys., 2015,142, 134305.

30 R. J. Doyle, Jr., J. Am. Chem. Soc., 1988, 110, 4120.31 A. P. Sergeeva, B. B. Averkiev, H. J. Zhai, A. I. Boldyrev and

L. S. Wang, J. Chem. Phys., 2011, 134, 224304.32 (a) M. Saunders, J. Comput. Chem., 2004, 25, 621; (b) P. P.

Bera, K. W. Sattelmeyer, M. Saunders and P. v. R. Schleyer,J. Phys. Chem. A, 2006, 110, 4287.

33 D. Yu. Zubarev and A. I. Boldyrev, Phys. Chem. Chem. Phys.,2008, 10, 5207.

34 (a) J. Chandrasekhar, E. D. Jemmis and P. v. R. Schleyer,Tetrahedron Lett., 1979, 20, 3707; (b) P. v. R. Schleyer,H. J. Jiao and M. N. Glukhovtsev, J. Am. Chem. Soc., 1994,116, 10129.

35 R. A. Kendall, T. H. Dunning, Jr. and R. J. Harrison, J. Chem.Phys., 1992, 96, 6796.

36 J. Cızek, Adv. Chem. Phys., 1969, 14, 35.37 G. E. Scuseria and H. F. Schaefer III, J. Chem. Phys., 1989,

90, 3700.38 R. J. Bartlett and M. Musial, Rev. Mod. Phys., 2007, 79, 291.39 E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E.

Carpenter, J. A. Bohmann, C. M. Morales and F. Weinhold,NBO 5.0, Theoretical Chemistry Institute, University ofWisconsin, Madison, 2001.

40 M. J. Frisch, et al., GAUSSIAN 09, Revision D.01, Gaussian,Inc., Wallingford, CT, 2009.

41 (a) P. Pyykko and M. Atsumi, Chem. – Eur. J., 2009,15, 12770; (b) P. Pyykko, J. Phys. Chem. A, 2015, 119, 2326.

42 A. Moezzi, R. A. Bartlett and P. P. Power, Angew. Chem., Int.Ed. Engl., 1992, 31, 1082.

43 A. Moezzi, M. M. Olmstead and P. P. Power, J. Am. Chem.Soc., 1992, 114, 2715.

44 G. Ferlat, T. Charpentier, A. P. Seitsonen, A. Takada,M. Lazzeri, L. Cormier, G. Calas and F. Mauri, Phys. Rev.Lett., 2008, 101, 065504.

45 J. D. Mackenzie, J. Phys. Chem., 1959, 63, 1875.

PCCP Paper

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article Online

29706 | Phys. Chem. Chem. Phys., 2015, 17, 29697--29706 This journal is© the Owner Societies 2015

46 The nature that the p sextet in B4O4+ (1) originates from

three O 2p lone-pairs is easily understandable. The B1/B3/B4centers only have 8 valence electrons after taking intoaccount the extra positive charge. Of these, one is used forthe terminal B–(BO) s bond and six participate in the six2c–2e B–O s bonds. The remaining one forms the deloca-lized 3c–1e s bond and renders s aromaticity for the system.Thus, the B1/B3/B4 centers cannot afford even a singleelectron for the p sextet. Indeed, orbital component analysesshow that O 2p atomic orbitals (AOs) constitute 90%, 95%,and 72% of the delocalized HOMO�5, HOMO�7, and

HOMO�9 orbitals, respectively (Fig. 3(a)). The AdNDP patternalso indicates the predominant role of O 2p AOs in the p sextet(Fig. 4(a)).

47 Despite the small NICS values in B4O4+ (1), the p sextet

appears to have a substantial bonding effect in terms ofenergetics. The completely bonding HOMO�9 (Fig. 3(a))possesses an orbital energy that is 2.39 eV lower thanHOMO�5, whereas HOMO�7 is only marginally lower thanHOMO�5 by 0.11 eV. This pattern also indicates that the Cs

distortion of B4O4+ (1) results in a relative minor splitting of

the HOMO�5/HOMO�7 orbitals.

Paper PCCP

Publ

ishe

d on

08

Oct

ober

201

5. D

ownl

oade

d by

Sha

nxi U

nive

rsity

on

19/0

4/20

17 1

4:31

:09.

View Article Online